From 5b976fc5da659812942073664b5f9ca8ec81f649 Mon Sep 17 00:00:00 2001
From: Joseph Ishac <jishac@nasa.gov>
Date: Mon, 7 Jun 2021 21:57:54 -0400
Subject: [PATCH] Making error handling when parsing messages consistent

---
 offlineimap/folder/IMAP.py    | 39 ++++++++++++++++++++++++-----------
 offlineimap/folder/Maildir.py |  9 +++++++-
 2 files changed, 35 insertions(+), 13 deletions(-)

diff --git a/offlineimap/folder/IMAP.py b/offlineimap/folder/IMAP.py
index 764f428..bfd232e 100644
--- a/offlineimap/folder/IMAP.py
+++ b/offlineimap/folder/IMAP.py
@@ -903,18 +903,33 @@ class IMAPFolder(BaseFolder):
         ndata0 = data[0][0].decode('utf-8')
         try: ndata1 = self.parser['8bit-RFC'].parsebytes(data[0][1])
         except:
-          e = exc_info()
-          response_type = type(data[0][1]).__name__
-          try: msg_id = \
-            re.search(b"message-id:.*(<[A-Za-z0-9!#$%&'*+-/=?^_`{}|~.@ ]+>)",
-            re.split(b'[\r]?\n[\r]?\n', bytes(data[0][1]))[0], re.IGNORECASE).group(1)
-          except AttributeError:
-            # No match
-            msg_id = b"<Unknown Msg-ID>"
-          raise OfflineImapError(
-            "Exception parsing message ({} type {}) from imaplib.\n {}: {}".format(
-              msg_id, response_type, e[0].__name__, e[1]),
-            OfflineImapError.ERROR.MESSAGE)
+            err = exc_info()
+            response_type = type(data[0][1]).__name__
+            msg_id = self._extract_message_id(data[0][1])[0].decode('ascii',errors='surrogateescape')
+            raise OfflineImapError(
+                "Exception parsing message with ID ({}) from imaplib (response type: {}).\n {}: {}".format(
+                    msg_id, response_type, err[0].__name__, err[1]),
+                OfflineImapError.ERROR.MESSAGE)
+        if len(ndata1.defects) > 0:
+            # We don't automatically apply fixes as to attempt to preserve the original message
+            self.ui.warn("UID {} has defects: {}".format(uids, ndata1.defects))
+            if any(isinstance(defect, NoBoundaryInMultipartDefect) for defect in ndata1.defects):
+                # (Hopefully) Rare defect from a broken client where multipart boundary is
+                # not properly quoted.  Attempt to solve by fixing the boundary and parsing
+                self.ui.warn(" ... applying multipart boundary fix.")
+                ndata1 = self.parser['8bit-RFC'].parsebytes(self._quote_boundary_fix(data[0][1]))
+            try:
+                # See if the defects after fixes are preventing us from obtaining bytes
+                _ = ndata1.as_bytes(policy=self.policy['8bit-RFC'])
+            except UnicodeEncodeError as err:
+                # Unknown issue which is causing failure of as_bytes()
+                msg_id = self.getmessageheader(ndata1, "message-id")
+                if msg_id is None:
+                    msg_id = '<Unknown Message-ID>'
+                raise OfflineImapError(
+                        "UID {} ({}) has defects preventing it from being processed!\n  {}: {}".format(
+                            uids, msg_id, type(err).__name__, err),
+                        OfflineImapError.ERROR.MESSAGE)
         ndata = [ndata0, ndata1]
 
         return ndata
diff --git a/offlineimap/folder/Maildir.py b/offlineimap/folder/Maildir.py
index 92ac124..3df4afa 100644
--- a/offlineimap/folder/Maildir.py
+++ b/offlineimap/folder/Maildir.py
@@ -262,7 +262,14 @@ class MaildirFolder(BaseFolder):
         fd = open(filepath, 'rb')
         _fd_bytes = fd.read()
         fd.close()
-        retval = self.parser['8bit'].parsebytes(_fd_bytes)
+        try: retval = self.parser['8bit'].parsebytes(_fd_bytes)
+        except:
+            err = exc_info()
+            msg_id = self._extract_message_id(_fd_bytes)[0].decode('ascii',errors='surrogateescape')
+            raise OfflineImapError(
+                "Exception parsing message with ID ({}) from file ({}).\n {}: {}".format(
+                    msg_id, filename, err[0].__name__, err[1]),
+                OfflineImapError.ERROR.MESSAGE)
         if len(retval.defects) > 0:
             # We don't automatically apply fixes as to attempt to preserve the original message
             self.ui.warn("UID {} has defects: {}".format(uid, retval.defects))